Development of the fingerprints for the quality evaluation of Viscum coloratum by high Performance liquid chromatography.

A high-Performance liquid chromatography coupled ultraviolet (HPLC-UV) method was developed for a chemical fingerprint analysis of Viscum coloratum. Eighteen peaks were selected as the common peaks and Homoeriodictyol-7-0-β-D-apiosiyl-(1→2)-β-D-glucoside was used as a reference. The relative areas of common peaks were used for hierarchical clustering analysis and similarity calculation. Thirty-seven samples collected from different sources were classified into five groups. The similarities of 21 batches Viscum coloratum samples were beyond 0.90. The results obtained suggest that the Chromatographie fingerprint can efficiently identify Viscum coloratum. Additionally, the fingerprints can then be used to evaluate the correlation between Viscum coloratum and hosts.

Introduction

Traditional Chinese medicines (TCMs) and their preparations have a long history in medical practice and health care, especially in Asia. Due to the effective therapeutic Performance, TCMs have attracted considerable attention in many fields. According to Chinese medicine theory, the whole components in crude herbs are responsible for the beneficial medicinal effects, which make the quality control of herbal produets very difficult. Traditionally, one or two markers or active components in herbs or herbal mixtures are used to assess the authenticity and quality of the com-plex TCMs. This strategy has been proved to be insufficient for the quality control of TCMs and their preparations because it does not evaluate all chemical components present in the Chromatographie profile. With the development of analytical technique, Chromatographie fingerprints have been widely used for the authentication and quality control of TCMs [1,2]. Chromatographie fingerprints techniques can be used to characterize both the marker compounds and the unknown components in a complex System, a strategy recommended to assess the quality and consistency of botanical produets by US Food and Drug Administration (FDA), the European Medicines Evaluation Agency (EMEA), and State Food and Drug Administration of China (SFDA) [3], [4], [5].

Viscum coloratum (Kom.) Nakai (V. coloratum) is a perennial evergreen, semi-parasitic plant which grows on branches or stems of deeiduous trees. It is known as Hujisheng in Chinese [6]. V. coloratum, an important medicinal herb suitable for commercial produetion, has been used for the treatment of various ailments including cardiovascular diseases, cancer, hepatitis and hemorrhage [7,8]. In clinical use, V. coloratum is usually mixed up with Taxillus chinensis (DC.) Danser. However, they belong to different genera of Loranthaceae. Hence, the development of the fingerprints is essential for efficiently identify V. coloratum. Otherwise, to grow and reproduce V. coloratum as a semiparasitic plant must successfully compete for a share of the host's water, avoid mineral deficiencies, tolerate differences in host xylem sap chemistry and, over time, flower and seed within the host canopy [9]. It acquires most of nutrients and water from their host through a root connection. So V. coloratum is restricted by the complicated interaction between V. coloratum and its host, and the impact of the host on the quality of V. coloratum is eritieal. So far, a few studies on fingerprints of V. coloratum using HPLC have been reported [10,11]. However, the small sample size limits to some extent the generalization of the findings made in the study. In addition, there has been no previous report on the evaluation of V. coloratum-host interactions. In the present study, a sensitive and aecurate HPLC fingerprints method was developed and validated for evaluating 37 batches of V. coloratum growing on different sources (host species not clear) and 5 batches of V. coloratum growing on different host species (the same sources). We aimed to provide a powerful way and some insight for the quality control of V. coloratum and further to find thc relationship between V. coloratum and its host.

Experimental

Chemicals and reagents

Acetonitrile was of HPLC grade (Richmond Hill, USA). All thc othcr reagents were of analytical grade from Concord Tech Reagent Company (Tianjin, China). Deionized water was uscd throughout the study.

The Standard Homoeriodictyol-7-O-β-D-apiosiyl-(1→2)-β-D-glucoside was isolated and purified by us from the branches and Steaks of V. coloratum. The structurc was charactcrized by chcmical and spectroscopic methods (UV, IR, NMR, MS) and then thc data wcre compared with those rcported by literature [12]. The purities of the isolated compound checked by HPLC-DAD at different wavclengths were above 98.9%.

Plant materials

Commercial herb samples of V. coloratum (host species were not clear, source group) were purchased from various pharmacies around China, while another V. coloratum (host species were clear, host group) were collected in Changbai Mountain (CBM, Yanji, Jilin). The collected plant material was dried at room tcmperature in the absence of light in a well-ventilated room. Professor Qi-Shi Sun authenticated the plant material, and the voucher specimens were deposited in the State Key Laboratory of Traditional Chinese Medicine (Shenyang Pharmaceutical University, China).

Apparate and Chromatographie conditions

The HPLC System was composed of a Shimadu LC-10ATvp series binary pump and a Shimadu SPD-10Avp UV detection. A Zhejiang University 2010 Chromatography Data System was used for data acquisition and integration. The analysis was carried out on a Phenomenex Synergi C18 column (250 mm×4.6 mm, 4 μm). The mobile phase was 0.5% glacial acetic aeid (A) and methanol-tetrahydrofuran (90:10) (B) with a gradient elution program of 3% – 20% (B) in 0–20 min, 20% – 70% (B) in 20 – 60 min and then re-equilibration of the column with 3% B for 10 min. The injeetion volume was 20 μL and the flow rate was maintained at 1.0 mL/min throughout thc elution. The column temperature was kept at 30 °C. The wavelength was set at 270 nm.

Preparation of Standard Solutions

In a clean, dry 10-mL Volumetrie flask, reference Standard Homoeriodictyol-7-O-β-D-apiosiyl-(1→2)-β-D-glucoside was accurately weighed and dissolved in methanol to make a reference Standard solution (20 μg/mL). The Standard solution was stored at 4 °C and brought to room temperature before use.

Preparation of samples

The V. coloratum samples were powdered in an electrical blender and then passed through a # 4 mesh sieve. Thc powder was accurately weighed (1.0 g) into a round-bottom flask and 25 mL of 50% (v/v) methanol was added. The flask was then weighed and heated in a water bath for 30 min. The flask was then cooled and 50% methanol was added in order to compensatc for the solvent loss. The extract was mixed well and transferred to a 50 mL centrifuge tube and centrifuged at 10000 rpm for 5 min. The supernatant was f iltered through a 0. 45 μm millipore f ilter membrane prior to injeetion.

Data processing

Hierarchical clustering analysis (HCA)

Clustcring is the art of grouping together pattern vectors that in some sense belong together because of similar characteristics. It provides both a visual representation of complex data and a method for measuring similarity between experiments. The similarity or dissimilarity between samples (objeets) is usually represented in a dendrogram for ease of interpretation. Each objeet is similar to thc others within one group but different from those in other groups with respect to a predetermined selection criterion [13]. In this part, source grouping of V. coloratum was performed using SPSS statistics Software (SPSS for Windows 16. 0, SPSS Inc., USA) based on the inter-group linkage method and squared Euclidean distance, respectively.

Similarity analysis

The Chromatographie profiles of all extracts were performed by Professional Software named Similarity Evaluation System for Chromatographie Fingerprint of TCMs, which was developed by Shenyang Pharmaceutical University (China). The Software was employed to evaluate the similarities between different chromatograms by calculating the cosine (C ) values of vectorial angle and correlation coefficient (r ) among samples [the calculation formula is described in Eq. (1) and Eq. (2)], as well as to computc and generate the mean chromatogram as a representative Standard fingerprint.where Xik and Xrk represent the kth variate of the ith sample and mean vector in common pattern, respectively; Xi and Xr are the mean values of them.

Results and discussion

Optimization of Chromatographie conditions

The Chromatographie conditions were optimized to obtain an ideal fingerprint containing sufficient information of constituents with good resolution of adjacent peaks within a short analysis time. Four types of Chromatographie columns, namely, Agilent Zorbax SB-C18 column, Phenomenex Luna C18 column, Phenomenex Synergi C18 column and Dikma Diamonsil C18, were chosen. The extract showed different retention behaviors on these columns. Phenomenex Synergi C18 column was more suitable for Separation of the constituents of the herb.

According to the UV absorption maxima of compounds on the three-dimensional chromatograms of HPLC-DAD detection, the monitoring wavelength was set at 270 nm. The Chromatographie peaks of extracts were identified by comparing the retention time of target compounds and UV spectrum with the chromatograms of Standards. Homoeriodictyol-7-O-β-D-apiosiyl-(1→2)-β-D-glucoside was used as a reference (Figure 1A). Relative retention time (RRT, the ratio between retention time of characteristic peaks to that of reference peak) and relative peak area (RPA, the ratio between peak area of characteristic peaks to that of reference peak) of each characteristic peak to reference were calculated in the chromatograms.

Figure 1

Chromatograms of Homoeriodictyol-7-O-β-D-apiosiyl-(1→2)-β-D-glucoside Standard (A) and Chromatographie fingerprint of V. colomtum from Mudanjiang (No.1) (B).

To obtain good Separation, methanol-water and acetonitrile-water were investigated as mobile phases. With acetonitrile-water the peaks of short retention time were not separated well. More compounds were separated well by use of methanol-water containing aeid and tetrahydrofuran. As a result, 0. 5% glacial acetic aeid (A) and methanol-tetrahydrofuran (90:10) (B) with a gradient elution program was chosen as the eluting solvent System. The chromatogram is shown in Figure IB.

Extraction procedure

Sample pretreatment conditions were also optimized by investigating the effect of extraction solvents and methods on the extraction efficiencies for different classes of chemical markers used for HPLC fingerprints. In the preliminary studies for the selection of extraction solvents (water, 25%, 50%, 75% and 100% methanol aqueous Solutions; 25%, 50%, 75% and 100% ethanol aqueous Solutions), 50% methanol aqueous solution was found to be the most effective solvent for extracting all compounds based on the HPLC results.

Extraction methods including ultrasonication and reflux were then investigated for extraction efficiencies (methanol as the extraction solvent). The HPLC peak areas of the compounds obtained from these two techniques were comparable, and heat reflux was chosen because of high extraction efficiency. Extraction time under reflux was also investigated, and the results showed that the most compounds were extracted within 30 min, and that longer period of reflux did not increase the contents significantly. The optimal extraction conditions for V. coloratum used in this study are presented in detail in Section 2.5.

Methods validation

The method was validated for parameters such as precision, repeatability and stability.

The precision was determined by replicate injection of the same sample solution (No.12) for five times. The relative retention time (RRT) and relative areas (RA) of common peaks were calculated. The relative Standard deviations (RSDs) of RRT and RA of common peaks for precision were found in the ränge of 0.1% –0.5% and 1.4% –5.1%, respectively.

The repeatability was determined by injection of five individual sample Solutions extracted from the same sample (No. 12) in the same way. RRT and RA of common peaks were calculated. The RSDs of RRT and RA were found in the ränge of 0.1% –0.2% and 1.3% –5.3%, respectively.

Stability was tested with one sample solution stored at room temperature for 0, 1, 3, 6, 12 and 24 h, and the areas of the main peaks were found to be stable within 24 h (RSD<5.0%). The RSDs of RRT and RA were found in the ränge of 0.1% –1.5% and 1.2% –4.0%, respectively.

All results indicated that the method was adequate, valid and satisfactory.

Establishment of Chromatographie fingerprint and similarity evaluation

Source group samples analysis

To standardize the fingerprint, 37 batches of samples (source group) were analyzed with the HPLC-UV analysis procedure. Peaks which existed in all batches of samples were assigned as common peaks. It was shown that there were 18 common peaks. Homoeriodictyol-7-O-β-D)-apiosiyl-(1→-2)-β-D-glucoside was chosen to calculate RRT and RPA. HCA procedure could find natural cluster of samples aecording to RPA. In this case, inter-group linkage was introduced and the squared Euclidean distance was calculated. HCA result is shown in Figure 2. Depending on the distance, it was clear that the source group samples could be classified into five clusters (Cluster I: 32 batches; Cluster II–V: 7 batches). The average chromatogram of the 32 batches of commercial samples was taken as the Standard characteristic fingerprint.

Figure 2

Results of hierarchical cluster analysis of 37 batches of Viscum coloratumsamples from different sources

SFDA suggests that all herbal chromatograms should be evaluated in terms of similarity by calculating the correlation coefficient and/or angle cosine value of original data. The results are listed in Table 1. The cosine and correlation coefficient values of 21 batches samples were more than 0.90 as Group I, 11 batches of samples ranged from 0.7 to 0.9 as Group II, and 5 batches of samples were less than 0.7 as Group III. Relatively higher similarity values indicated the consistency of V. coloratum which ensured the stability of preparations. Therefore, if 0.90 is set as an appropriate threshold, it is easy to identify V. coloratum based on the Chromatographie fingerprint. The fingerprint patterns of different samples are different, and by comparing each fingerprint pattern with the mean chromatogram, we can obtain the similarity value of each fingerprint pattern, which can help us evaluate the quality of different samples.

Table 1

Evaluation of the similarity of V. colomtum from different origins and different host species

		Similarity
No.	Origin	Cosine	Correlation coefficient
1	Mudanjiang, Heilongjiang	0.99	0.99
2	Huhehaote, Inner Mongolia	0.78	0.71
3	Tangshan, Hebei	0.99	0.99
4	Weinan, Shaanxi	0.94	0.92
5	Nanyang, Henan	0.99	0.99
6	Lanzhou, Gansu	0.98	0.97
7	Haerbin, Heilongjiang	0.99	0.99
8	Wulumuqi, Xinjiang	0.98	0.97
9	Taiyuan, Shanxi	0.86	0.80
10	Yanji, Jilin	0.99	0.99
11	Bozhou, Anhui	0.99	1.00
12	Pingyao, Shanxi	0.84	0.76
13	Nanjing, Jiangsu	0.96	0.94
14	Dongping, Shandong	0.85	0.79
15	Zhongsheng, Shandong	0.98	0.98
16	Zhongsheng, Shandong	0.98	0.98
17	Dongying, Shandong	0.85	0.78
18	Anshan, Liaoning	0.97	0.96
19	Wulian, Shandong	0.89	0.86
20	Shenyang, Liaoning	0.99	0.98
21	Zhongsheng, Shandong	0.97	0.97
22	Zhongsheng, Shandong	0.97	0.95
23	Zhongsheng, Shandong	0.91	0.87
24	Hangzhou, Zhejiang	0.97	0.96
25	Shenyang, Liaoning	0.88	0.83
26	Kunming, Yunnan	0.57	0.44
27	Shanghai	0.90	0.86
28	Shijiazhuang, Hebei	0.99	0.98
29	Chengdu, Sichuan	0.98	0.99
30	Daqing, Heilongjiang	0.54	0.40
31	Kulun, Inner Mongolia	0.96	0.95
32	Yingkou, Liaoning	0.68	0.52
33	Bin county, Heilongjiang	0.96	0.95
34	Kunming, Yunnan	0.43	0.28
35	Bin county, Heilongjiang	0.84	0.76
36	Anshan, Liaoning	0.51	0.37
37	Guilin, Guangxi	0.87	0.81
Hl	Populus ussuriensis Kom.	0.93	0.94
H2	Tilia amurensis Rupr.	0.91	0.89
H3	Betula platyphylla Suk.	0.91	0.91
H4	Ulmus pumila L.	0.85	0.76
H5	Pyrus ussuriensis Maxim	0.92	0.89

Host group samples analysis

The correlation between V. coloratum and its host was also investigated. The chromatogram of V. coloratum showed drastic differences from that of the host by direct visual inspection (Figure 3). So in order to distinguish V. coloratum from different hosts, the similarities between the entire Chromatographie profiles of 5 batches of V. coloratum (hosts were clear) and the Standard characteristic fingerprint were evaluated. The results suggested that the fingerprints of the V. coloratum samples from different hosts were similar (Table 1). The relationship among the V. coloratum and their host species needs to be clarified. Further studies are currently in progress in our laboratory.

Figure 3

Chromatographie fingerprints obtained from 5 batches of V. coloratum from different hosts

Conclusion

A simple, reliable and aecurate HPLC-UV method has been developed for the fingerprint analysis of the major compounds in V. coloratum. The fingerprint of the V. coloratum sample showing eighteen common peaks represents the characteristics of the herb's constituents and provides an aecurate method for the quality control of V. coloratum. HCA suggested that 37 batches of source group samples could be divided into five groups, and it revealed that there was a considerable Variation between chemical constituents and plant populations. Meanwhile, the cosine and correlation coefficient values of 21 batches of samples were more than 0.90. If 0. 90 is set as an appropriate threshold, it is easy to identify V. colomtum based on the Chromatographie fingerprint. Relatively higher similarity values indicated the consistency of V. coloratum which ensures stability of preparations.